search

clab / lstm-parser

Transition-based dependency parser based on stack LSTMs
Apache License 2.0
204 stars 63 forks source link

cnn is out of memory #10

Closed HenrikLovold closed 7 years ago

HenrikLovold commented 7 years ago

When attempting to train a model, the program aborts with the report

cnn is out of memory, try increasing with --cnn-mem

However, this argument is not accepted by the jar. I don't know how to fix this. Any help would be great!

miguelballesteros commented 7 years ago

Hi Henrik,

You do not need to add this argument when running the jar file. The jar file should be run exactly in the same way as in the readme (it just generates the data format). You may increase the memory when running the parser with --cnn-mem:

parser/lstm-parse --cnn-mem 4096 -T trainingOracle.txt -d devOracle.txt --hidden_dim 100 --lstm_input_dim 100 -w sskip.100.vectors --pretrained_dim 100 --rel_dim 20 --action_dim 20 -t -P